key: cord-0924227-1gr2a9c7
authors: Arjmand, Sareh; Hosseinkhan, Nazanin
title: Identification of potential coagulation pathway abnormalities in SARS-Cov-2 infection; insights from bioinformatics analysis
date: 2020-12-07
journal: bioRxiv
DOI: 10.1101/2020.12.07.414631
sha: 88782f2e0146855a11dfa25708fa9fdc982c7cae
doc_id: 924227
cord_uid: 1gr2a9c7

Abnormal coagulation parameters have been explored in a significant number of severe COVID-19 patients, linked to poor prognosis and increased risk of organ failure. Here, to uncover the potential abnormalities in coagulation pathways, we analyzed the RNA-seq data (GEO147507) obtained from the treatment of three pulmonary epithelial cell lines with SARS-CoV-2. The significant differentially expressed genes (DEGs) were subjected to Enrichr database for KEGG pathway enrichment analysis and gene ontology (GO) functional annotation. The STRING database was used to generate PPI networks for identified DEGs. We found three upregulated procoagulant genes (SERPINE1, SERPINA5, and SERPINB2) belong to the serine protease inhibitor (serpin) superfamily that inhibit tissue plasminogen activator (t-PA) and urokinase plasminogen activator (u-PA) in the fibrinolysis process. In conclusion, we suggest the fibrinolysis process, especially the blockage of t-PA and u-PA inhibitors, a potential target for more study in treating coagulopathy in severe COVID-19 cases.

The current crisis that emerged due to the rapid national and international spread of the new pathogenic SARS coronavirus (SARS-CoV-2) has posed tremendous challenges to the global public health and economy. The number of positive cases and daily deaths is exhibiting a constant progression. Many scientists from different disciplines collaborate to identify possible interventions to reduce the spreading rate and seek effective therapeutics targets. Knowledge obtained from the SARS-CoV (the cause of the previous SARS epidemic) greatly helped to understand the new coronavirus and accelerated its research.

Many of the common and rare clinical features of COVID-19, such as fever, dry cough, myalgia, fatigue, shortness of breath, diarrhea, Acute Respiratory Distress Syndrome (ARDS), vomiting, liver damage, and septic shock, have been well-considered to categorize and monitor disease progression to improve the clinical outcomes 1, 2 .

After infection, SARS-COV-2 triggers immune responses that, in turn, results in localized lung inflammation. However, in some patients, uncontrolled cytokine responses with unknown causes lead to hyper-inflammation, described as a cytokine storm, and a considerably higher risk of multiorgan failure and death 3 . A broad spectrum of anti-cytokine and anti-inflammatory therapeutic strategies have been considered to improve patients' clinical outcomes and prognosis. These immunomodulators must be used in the right time and balanced manner to maintain a sufficient inflammatory response for virus clearance 4, 5 . However, the hyper-inflammatory state is not the sole factor responsible for lung injuries and organ failure. Rather, the cross-talk between inflammation and coagulation pathways, through which activation of one system amplifies activation of the other, may lead to substantial tissue damage and organ failure in critically ill patients 6 . Though some studies are starting to show the importance of associated coagulopathy in the prognosis of COVID-19 patients, few clear knowledge has been provided on the hemostasis dysfunction mechanisms in them. Furthermore, inhibition of the coagulation pathway, as a therapeutic target, received less attention 7 .

Blood coagulation is a dynamic process, triggered as a cascade of proteolytic reaction in which each proenzyme is converted to its active form by an upstream-activated coagulation factor. In the central step of coagulation, Factor Xa converts prothrombin (Factor II) into the active thrombin (Factor IIa). Thrombin works as a master protease that converts the soluble fibrinogen (Factor I) 4 to fibrin (Factor Ia) to form the clots. Thrombin also converts Factor XIII to its activated form (Factor XIIIa), which renders the fibrin monomers to stable cross-linked networks 8, 9 . Following clot formation, a separate process called fibrinolysis occurs to degrade the fibrin, restore blood vessel patency, and prevent possible thromboembolism. Plasmin is the main proteolytic enzyme that breakdown the fibrin mesh, leading to the production of circulation fragments (fibrin degradation products (FDP) and D-dimer) cleared by other proteases or by the kidney and liver. Plasminogen is the inactive precursor of plasmin that circulates in free form and is activated, after binding to fibrin, by t-PA and u-PA 10 . Impaired fibrinolysis may result in a delayed breakdown of clots and predisposes to thromboembolism 11 . The scheme of coagulation and fibrinolysis is shown in Figure 1 . Whole-transcriptome RNA-seq analysis is a high-throughput approach for precise measurement of the expression levels of transcripts 12 . This method can give an objective view of the molecular mechanism underlying the virus pathogenesis and interaction between host and virus.

The present study aimed to analyze the RNA-seq data (GEO147507) that was provided by Tenoever et al. from primary human lung epithelium (NHBE), transformed lung alveolar (A549) cells, and transformed lung-derived Calu-3 cells infected with SARS-Cov-2 (USA-WA1/2020).

DEGs were identified in infected cells 24h post-infection. We compared outcomes with the obtained results from the infection of cells with Avian Influenza Virus (AVI) subtype H1N1, the bird flu with pandemic potential for humans. We analyzed the dataset to uncover the potential abnormalities in coagulation pathways that occur due to the infection with SARS-CoV-2. We hope that this study offers new insights into the potential mechanisms involved in SARS-CoV-2 pathogenesis and would prove a useful theoretical basis for future biological and clinical applications.

Transcriptional response to SARS-CoV-2 infection dataset GSE147507 submitted by Tenoever BR., and Blanco-Melo D. based on GPL18573 platform (Illumina NextSeq 500) was downloaded from GEO database, in the National Center for Biotechnology Information (NCBI).

6 FASTQ files were introduced to FASTQC algorithm to check the presence and the type of adaptor contaminations within FASTQ files. Trimming of FASTQ files was then carried out using Trimmomatic algorithm. We used GRCH37 (hg19) version of the reference genome for mapping reads using "Bowtie for Illumina" algorithm. Read counts for each feature (gene) were then calculated using "HTSeq-count" algorithm. Finally, we employed "DESeq2" algorithm on the resulted read counts to get the list of deregulated mRNAs. Applying the adjusted p-value ≤0.05, the most statistically significant up and down-regulated genes were selected. To identify deregulated genes, the biologically significant changes between mock and treated samples, |logFC| >= ±0.58 cutoff was applied, which equals to at least 1.5 folds increase or decrease in expression values.

The significant DEGs were mapped into Enrichr database for KEGG pathway enrichment analysis and GO functional annotation. A p < 0.05 depicts significant enrichment.

The STRING database was used to predict the interaction relationship between DEGs corresponding to coagulation. Only the interactions with experimental support were chosen for PPI data acquisition. The reliability of PPI interactions were checked based on the provided experimental information. The integrated PPI network was visualized and analyzed using cytoscape v3.8.2.

The list of DEGs in all cell types treated with SARS-CoV-2 and H1N1 is provided in supplementary 1. The identified DEGs enriched in the coagulation pathway are shown in Table 1 . 

The identification of proteins that interact directly with DEGs involved in the coagulation pathway could help understand the molecular mechanism of hypercoagulation in COVID-19

patients. In the present study, two PPI networks were built for the results of SARS-CoV-2 (three upregulated genes) and H1N1 (three upregulated and one downregulated genes) separately and compared ( Figure 2 ). 

Based on the severity of the symptoms, patients with COVID-19 is classified into four levels;

mild, moderate, severe, and critical. While mild patients, which constitute a large number of patients, exhibit weak or no symptoms, the critical patients quickly progress to destructive symptoms such as ARDS, coagulation dysfunction, multiple organ failure, and loss of consciousness 2 .

9 A significant number of cases with severe COVID-19 infection experience clotting problems that were more likely to receive ICU admission. It is found that blood clots are closely linked to the increased risk of organ failure and death 4 .

Coagulation is an integral part of innate immune responses that slows down the circulation of invasive microorganisms and limits immune components' loss. In turn, the natural anticoagulants modulate their function to limit excessive coagulation, maintain the hemostatic balance, and reduce tissue destructions 13 Coagulation dysfunction has been proposed as the complications of COVID-19, which can ultimately lead to ischemic stroke, myocardial injury, and pulmonary thromboembolism, particularly in individuals with underlying comorbid diseases [18] [19] [20] [21] .

The site of thrombosis in the majority of severe COVID-19 is the lungs 22 . Postmortem examination of lungs in COVID-19 patients has revealed widespread microthrombotic occlusion of small pulmonary vessels 23, 24 . The incidence of pulmonary embolism has been measured by 23-30% in severe COVID-19 patients 25 . The pulmonary thromboembolism (PTE) is diagnosed in hospitalized COVID-19 patients by observation of the sudden onset of oxygenation deterioration and shock 26 . It seems that a variety of potential risk factors, including infection, the immobilization of patients due to prolonged ICU admission, mechanical ventilation, hypoxia, and the hypercoagulable state predisposes them to the high risk of PTE 27, 28 .

Early detection of life-threatening blood clots in the lung could lead to rapid treatment intervention and reduce the mortality rate. However, a specific diagnosis of PTE is challenging.

Understanding the molecular mechanism of SARS-CoV2 behavior and host response to the virus SERPINE1 gene encodes the plasminogen activator inhibitor 1 (PAI-1), which is the most important inhibitor of t-PA and u-PA in the fibrinolysis process. PAI-1 is secreted as an active protein and, by binding to t-PA and u-PA, blocks the activation of plasminogen to plasmin, and therefore the fibrin clot hydrolysis. It was indicated that the production and secretion of PAI-1 could be induced by stimulators like thrombin, endotoxin, cytokines, and chronic inflammation 35, 36 . This inhibitor is required for the downregulation of fibrinolysis and the control of blood clot degradation 37 . High levels of circulating PAI-1 was shown to be associated with some thrombotic diseases and cancers 38, 39 , and transgenic mice overexpressing native human PAI-1 developed spontaneous thrombosis 40 . Overexpression PAI-1 has been known as a biomarker of agingassociated thrombosis and a variety of other pathologies (e.g., obesity, hyperinsulinemia, type-2 diabetes, coronary heart diseases, a decrease of immune responses, a proliferation of inflammation, and vascular sclerosis/remodeling) 41, 42 . Based on this information, it seems that this inhibitor could be considered a notable target for inhibition in COVID-19 patients at the risk of developing pulmonary thromboembolism. It is noteworthy that already some fibrinolytic drugs, such as heparin and tPA, are administering to remedy the thrombotic pathology in COVID-19, and it is suggested this therapy is effective in some patients 22, 43 . Suppressing of PAI-1 (using specific antibodies, antisense oligonucleotides, peptide antagonists, etc.) has been suggested for further studies as a potential therapy for complications such as renal fibrogenesis, breast cancer, and vascular diseases 39, 41, 44 . Furthermore, a wide range of small molecules with anti-PAI-1 activity have been isolated from natural sources that showed satisfactory results in vitro and with some in vivo models 45 . Interestingly, the treatment of A549 with the H1N1 AVI virus (with higher MOI (=5) compared to the SARS-CoV-2) showed completely different results. The analysis indicated that AVI induces the expression of two anticoagulants gene (plasminogen activator urokinase receptor (PLAUR), and tissue factor pathway inhibitor (TFPI)) and downregulation of protein C (PROC), the zymogen of activated protein C (APC).

The results of the analysis on the other bronchial epithelial cell (NHBE) with the same MOI (=2) indicated the ~2.24 fold (logFC=1.24) overexpression of SERPINB2 that is involved in the coagulation pathway. SERPINB2 or plasminogen activator inhibitor 2 (PAI-2) is a well-described coagulant factor that inactivates u-PA and t-PA, and impaired fibrinolysis. It has been shown that the expression of PAI-2 is associated with placental tissue and monocyte macrophages 46 . PAI-2 level is normally very low in blood but is drastically elevated during pregnancy, partially explaining the higher risk of thrombosis during pregnancy 47 . It is also markedly induced by proinflammatory mediators, such as tumor necrosis factor-alpha (TNF-α) and LPS, and a variety of cytokines and growth factors that are involved in cell differentiation 48 Treatment of Calu3, another lung epithelial cell used in this study, with SARS-CoV-2 led to a significant increase (~1.3 fold, LogFc=1.63) in serpinA5 or plasminogen activator inhibitor 3 (PAI-3), known as protein C inhibitor (PCI) as well. PAI-3 is a multi-functional serpin playing several roles depending on organs, tissues, and species 53 . This protein affects blood coagulation and fibrinolysis in different ways. Initially, it was found to be the inhibitor of activated protein C 14 (APC). APC is a natural anticoagulant that down-regulates the blood coagulation by proteolytic degradation of FVa and FVIIIa 54 . PAI-3 also inhibits u-PA, t-PA, and several other serine proteases involved in blood coagulation and fibrinolysis 55 .

Overall, here we found that SARS-CoV-2 treatment led to the upregulation of three procoagulant genes in pulmonary epithelial cells. Interestingly, the three upregulated genes' common feature is their involvement in the fibrinolysis process and inhibition of clot degradation.

As it is clear from the PPI network, all three detected serpin interact with u-PA (in Figure 2 is shown as PLAU), and two of them, SERPINE1 and SERPINA5, was induced by thrombin. The equal experiment using H1N1 viruses that affects the pulmonary system as well led to contradictory results, and the detected upregulated genes have known for their anticoagulatory effects through the inhibition of the coagulation pathway and inhibition of clot formation. From the result of PPI for H1N1 it seems that this virus does not affect fibrinolysis directly.

In conclusion, we suggest the fibrinolysis process, especially the blockage of t-PA and u-PA inhibitors, a considerable and potential target for preventing and treating coagulopathy in severe COVID-19 cases.

Epidemiology and clinical features of COVID-19: A review of current literature

Unique epidemiological and clinical features of the emerging 2019 novel coronavirus pneumonia (COVID-19) implicate special control measures

Understanding SARS-CoV-2-mediated inflammatory responses: from mechanisms to potential therapeutic tools

COVID-19 cytokine storm: the interplay between inflammation and coagulation. The Lancet Respiratory Medicine

Weathering the cytokine storm in COVID-19: therapeutic implications

Cross talk pathways between coagulation and inflammation

Prominent changes in blood coagulation of patients with SARS-CoV-2 infection

Structural-functional relationships in the human thrombin A-chain: University of British Columbia

The blood coagulation cascade

Molecular mechanisms of fibrinolysis

Decreased plasma fibrinolytic potential as a risk for venous and arterial thrombosis. Seminars in thrombosis and hemostasis

RNA-Seq: a revolutionary tool for transcriptomics

Innate immunity and coagulation

PAMPs and DAMPs as triggers for DIC

Abnormal coagulation parameters are associated with poor prognosis in patients with novel coronavirus pneumonia

Prevalence of venous thromboembolism in patients with severe novel coronavirus pneumonia

Multiple embolic stroke on magnetic resonance imaging brain in a COVID-19 case with persistent encephalopathy

Association of coronavirus disease 2019 (COVID-19) with myocardial injury and mortality

Acute COVID-19 and the Incidence of Ischemic Stroke and Acute Myocardial Infarction

Pulmonary embolism in patients with coronavirus disease-2019 (COVID-19) pneumonia: a narrative review

Perspective on fibrinolytic therapy in COVID-19: the potential of inhalation therapy against suppressed-fibrinolytic-type DIC

Pulmonary vascular endothelialitis, thrombosis, and angiogenesis in Covid-19

Autopsy findings in COVID-19-related deaths: a literature review

Acute pulmonary embolism associated with COVID-19 pneumonia detected by pulmonary CT angiography

Pulmonary embolism or pulmonary thrombosis in COVID-19? Is the recommendation to use high-dose heparin for thromboprophylaxis justified?

Diagnosing pulmonary thromboembolism in COVID-19: a stepwise clinical and imaging approach

Attention should be paid to venous thromboembolism prophylaxis in the management of COVID-19

Imbalanced host response to SARS-CoV-2 drives development of COVID-19

Reactive centre loop dynamics and serpin specificity

Primary Human Lung Alveolus-on-a-chip Model of Intravascular Thrombosis for Assessment of Therapeutics

The alveolar epithelium can initiate the extrinsic coagulation cascade through expression of tissue factor

Inhibition of PAI-1: a new anti-thrombotic approach

A Noncanonical Role for Plasminogen Activator Inhibitor Type 1 in Obesity-Induced Diabetes

Plasminogen activator inhibitor type-1 as a regulator of fibrosis

Plasminogen activator inhibitor-1 in cancer research

Inhibition of PAI-1 expression in breast cancer carcinoma cells by siRNA at nanomolar range

Development of venous occlusions in mice transgenic for the plasminogen activator inhibitor-1 gene

Inhibition of PAI-1 antiproteolytic activity against tPA by RNA aptamers

Plasminogen activator inhibitor-1 is a marker and a mediator of senescence

Fibrinolytic therapy for refractory COVID-19 Acute Respiratory Distress Syndrome: Scientific rationale and review

Plasminogen activator inhibitor type 1 is a potential target in renal fibrogenesis

Small molecules inhibitors of plasminogen activator inhibitor-1-An overview

Purification and characterisation of plasminogen activator inhibitor 2 produced in Saccharomyces cerevisiae

Elevated procoagulant endothelial and tissue factor expressing microparticles in women with recurrent pregnancy loss

Picornavirus receptor downregulation by plasminogen activator inhibitor type 2

Hypersensitivity of aryl hydrocarbon receptor-deficient mice to lipopolysaccharide-induced septic shock

Local abnormalities in coagulation and fibrinolytic pathways predispose to alveolar fibrin deposition in the adult respiratory distress syndrome

Expression of plasminogen activator inhibitors type-1 and type-2 in the mouse lung after administration of crystalline silica

C1-inhibitor efficiently delays clot development in normal human whole blood and inhibits Escherichia coli-induced coagulation measured by thromboelastometry

The multi-functional serpin, protein C inhibitor: beyond thrombosis and hemostasis

The protein C pathway

Protein C inhibitor (PAI-3): structure and multi-function